This paper presents a method of optimum design of multilayer-antireflection films based on database of material properties. In the design process of the optical multilayer films, the designer must determine appropriate refractive index and thickness of each thin film in accordance with design specifications on reflectance distribution. There lies the difficulty of the design where it is not clear such as the formulation of the multilayer structure to minimize the reflectance at the prescribed wavelength range. Furthermore, even if the refractive index is determined it does not always follow that the material which has the specified refractive index is available. To overcome these difficulties, the genetic algorithm and the database are used for the optimization process. The way to apply the genetic algorithm to design the structure of the multilayer films and its optimization are described in detail. Some results of typical problems are shown to demonstrate and discuss the effectiveness of the proposed method.
Authors of this paper constructed a remote control simulation system of micromachine, to make clear experimentally the accomplishment of guiding tasks and the formation of man's cognitive models of space configuration when man remote-controls a micromachine in some fine tubes. The system gives pictures of the inner part of the test space. Subjects were set to a task to guide a micromachine from a start point to a goal point by relying on these 'subjective' visual images. The test space was set to 2-dimensions and motions of the micromachine were restricted to going forward-and-backward and turning right-and-left. The authors had the subjects draw objective space figures after trials to check their cognitive models in their brain. Results show that (1) subjects who were given only subjective visual images settle their own paths to the goal, and form partial cognitive models along the settled paths. Their times required to reach the goal and strokes become stable. (2) Having subjects draw a figure after every trial stimulates formation processes of their global models of the test space. (3) Familiar marks or symbols inside tubes help subjects to form better cognitive models, and make them accomplish guiding tasks more easily.
The basic technology to measure rapidly the surface roughness of 3-dimensional cast metals using an optical non-contact method has been developed. In the experiment, the standard plates with surface roughness ranging from 100 μm to 400 μm for the evaluation of cast metal has been used. The standard plate is illuminated obliquely with the incident angle θi (=60°) and detected by CCD camera perpendicularly. An evaluated value F is calculated using the detected image. F is proportional to the surface roughness Ry when the standard plate is illuminated from both side at a right angle to inclined direction, and the plate is inclined with angles ranging from +25° to -25°. The surface roughness Ryis obtained by F with the error less than ±18% according to the proportional relationship between F and Ry.
The profile measurement system using time sequential space encoding with multi-gray code patterns has been developed. However, some objects which have colored surfaces and the unevenness of reflectivity could not be measured using this method. Therefore the reference projection images are added to the system. By the reference projection image method, the colored plaster figures under the normal illumination could be measured by comparing objected space coding images with the reference images. However, the number of projections increased from that of the conventional binary space code projection method. In this paper, the novel space code encoding system, the self-referenced pattern projection method, was introduced. This method reduces the number of projections and keeps the features of the previous method. Using this system, the measuring space is coded into n! spaces with the n projections which has the n gray levels. It is shown that the number of projection of this method is the least in normal projection situations.
The self-calibration method which detects the derivative function of the calibration curve without using any references or any additional sensors has been applied to an interferometric microscope. Two specimens were used to get the discrete derivative function of the calibration curve. One is an inclined flat surface which gives a continuous input value of the calibration. The difference between two measurements which are obtained before and after a small shift in the height direction is used for getting a derivative function of the wanted calibration curve. The other is a step specimen whose stepheight determines the small displacement of the calibration curve to get discrete derivative of the function. The results by these two methods have shown good agreement with each other to confirm the effectiveness and reliability of the proposed method. With the calibration and the correction, the variation of small step-height measurement is reduced to 0.5% which is 1/4of the original one.
This paper describes an analyzing method of a tape stiffness which is fixed with tape-head contact performance at the tape-edge. New magnetic tape systems have been using the narrow-width tape and the nearby area of the tape-edge to increase an efficiency of recording area of tape, therefore, the analysis of tape stiffness in the width direction has been very important approach to the high density recording. A new method by the laser measures deformation of the tape that is pressed to the precision sphere on the drum, and estimates the tape stiffness by the deformation. The results of an experiment coincide well to the calculative deformation, and reappear well. This analyzing method settles the tape stiffnesss in the width direction of narrow tape to obtain the best tape-head contact in and around the tape-edge. The tape having the most suitable stiffness contributes to the high density recording system.
The viewing angle used in a previously developed laser-beam pipe-scanning sensor has been optimized to improve measurement accuracy. The viewing angle is the angle between the optical paths-that of the beam from the laser diode (LD) and that of the light scattered to the viewing lens. The trade-off between the scattering characteristics of the measured surface and the magnifying power of the viewing lens was then used to determine the best viewing angle. The results of this method agreed with the angles determined experimentally by manufacturers of displacement meters. In the previous sensor, the laser power was constant. However, the reflection ratio on the inner wall of a conduit varies substantially, so the incident light scattered from the surface to the position sensor vanes significantly. The output signal thus becomes saturated, and the S/N ratio is degraded, making it impossible to measure the radii accurately. It is now changed appropriately when the reflectivity of the scanned pipe wall changes by using analog feedback control circuits, in which the laser power signal and the output signal of the position sensor are fed back in parallel. With these two improvements, the laser-beam pipe-scanning sensor can more accurately measure the inner radii of conduits.
The hardness of alumina ceramic cutting tool can be increased about 1.5-1.6 times by putting titanium oxide. The alumina ceramic cutting tool based on purified alumina oxide material is embedded in the titanium oxide powder, and then is thermally treated in an electric furnace to diffuse the oxide additive from the surface to inside by the solid solution. After adding titanium oxide, the Vickers hardness of the surface layer of the ceramic cutting tool is increased from the original 1800HV0.5 to 2900HV0.5. The result is that after titanium oxide additive treatment, alumina ceramic cutting tool has distinctly the restrictive effect on flank wear and crater wear in cutting process. But when the temperature of the thermal treatment is too high, the wear of the cutting tool occurs very rapid, because the alumina grains of the alumina ceramic cutting tool become bigger.
Cutting performance and wear mechanism of PcBN tools in the finishing operation of gray cast iron (FC300) are investigated. The tool life of the case using the high cutting speed condition exceeding 16.7m/s is improved remarkably as compared with the case of usual condition at 6.7m/s. The wear development rate at the condition of cutting speed of 25.0m/s is relatively lower than the other conditions. In order to clarify themain cause of wear at high speed cutting condition, adhesive formed on the tool flank face is observed in detail. It was confirmed that wear behavior of the tool can be classified by the cutting speed of 25.0m/s. At the condition of cutting speed of 25.0m/s, the adhesion on tool surface is composed by aluminum oxide. And the reduction of flank wear rate is caused by the working of aluminum oxide as a protective film. On the other hand, at the condition of cutting speed up to 25.0m/s, the Si is detected in adhesion on flank face. The formation of the SiAlON phase as the results of reaction between Al compound in the tool binding phase and Si in the workpiece material is observed.
As the second report of serial studies on welding by high energy beams, penetration mechanisms by carbon dioxide gas laser for steels were discussed. On the basis of systematic bead-on-plate tests, the fundamental welding characteristics of laser beam were estimated in comparison with that of an electron beam. The results obtained are as follows : (1) As an inevitable consequence that the laser beam in the key hole is mainly absorbed at the side wall, input power density to the bottom and aspect ratio of penetration by laser beam are lower than those by an electron beam. A depth and an opening of the key hole are in the state of equilibrium at any welding speed, then, the aspect ratio is almost constant independently of the speed. (2) The welding speed for the laser is not restricted at higher side by an ability of penetration which declines in proportion to the involution of speed but by an outbreak of humping bead, similar to the electron beam. (3) The penetration formation mechanisms for steels of a laser and an electron beam can be treated equally by using the wedge type volume source model based on the heat conduction theory.
This paper describes the measurement of the change in the absorption factor with extension of the depth of the hole pierced with a laser beam by means of the inverse problem solution. First, the temperature-rise at a certain point on a workpiece is measured by a thermocouple when a focused laser beam is irradiated on the workpiece for a preset pulse duration t1. Assuming that the absorption factor is η1, the temperature-rise is calculated by the finite-difference analysis. The calculation is repeated until the η1, which produces coincidence between the calculated and measured temperatures is obtained. The workpiece is then replaced with a new one and the temperature-rise is measured for a pulse duration t2=t1+Δt. Assuming that the absorption factor for t1<t<t2 is η2, the calculation of the temperature change at the point is iterated until η2, where the calculated and measured temperature changes coincide at the same point, is obtained. Repeating this process, the change of the absorption factor with the increase of the depth of the processed hole can be determined. It was found that the absorption factor increases with increasing the depth of the processed hole and reaches 96% at the moment when penetration occurs, but subsequently it decreases rapidly after the penetration. When oxygen is used as the process gas in place of nitrogen, the maximum absorption factor reaches 133% of the incident laser power as a result of the generation of heat by the oxidation of the workpiece material.
This study deals with the optimization of ceramics grinding operation with metal bonded diamond wheels.In this paper, the discrimination of the quality of wheel conditioning and of dressing interval is discussed based on signals obtained from an accelerometer pickup mounted on the spindle head of the surface grinder. As a result, it is clarified to identify whether truing and dressing operations are completed or not, by detecting the specific frequencies which are equivalent to the integral multiple numbers of spindle revolution and the principal characteristic frequency of the spindle head, respectively. Also, with 800Hz which is the characteristic frequency of the spindle head, as the specific frequency, it will be possible to determine the dressing interval by considering the rapid increase of acceleration amplitude.
This paper describes aspheric mirror fabrication technology for an imaging optics of EUV (Extreme Ultraviolet) lithography, which is a candidate of mass-production technology for 0.1-μm design rule ICs. In this optics, accuracy of aspherical mirrors must be sub-nm-order to provide good imaging performance. To clarify technical problems, a multi-tool polishing system for a quarter-size prototype imaging optics were developed. These tools' polishing loads are dynamically controlled to correct local figure error. As a result, 3.4 nm rms figure precision was obtained for a concave aspherical mirror. Furthermore, 0.1-μm line & space patterns were partially replicated in a 1.4 mm-wide ring-field. A major remained problem to the goal is how to decrease mid-frequency figure error caused by tool movement.
Fuzzy control is applied to the superfinishing in order to optimize the cutting ability of the stone. The finishing performance of vitrified bonded WA stone is regarded as the fuzzy goal. At the beginning of cutting, the stone is in the stage of cutting to get high stock removal rate. But the cutting ability is decreasing as the finishing proceeds and is lost at the end of finishing, resulting in getting the smooth surface. A cutting resistance is used to monitor the cutting condition of stone. The cutting ability of stone is controlled by both the stone pressure and the amount of electricity for electrolytic in-process dressing. A water-solution type coolant and metal bonded diamond stone are used. The main results obtained are as follows. Fuzzy control makes it possible to give the optimized cutting ability to the diamond stone. In the first case, it is possible to get the large stock removal following the smooth surface in a limited time. In second case, it is possible to minimize the finishing time in which the finishing performance intended in stock removal and surface roughness is obtained. In third case, it is possible to keep the cutting ability of stone constant for a long time.
Lapping performance is essentially attributed to cutting abilities of grain, which depends on the grain shape. However, it was very difficult to evaluate the grain shape quantitatively, because the grain shape was changing complicatly during lapping. In this study, it was examined how the grain shape characteristics change with lapping pressure. In order to evaluate the grain shape, a new general attempt was made by the distribution of grain size and the aspect ratio from macroscopic point of view, by the fractal dimension from microscopic point of view. Here, the term “fractal dimension” can be defined as the complication of grain external line. As a result, macroscopic and microscopic characteristics of grain shape varied with lapping pressure. Increasing lapping pressure tended to be less the grain size and the aspect ratio, higher the fractul dimension. Namely, it follows that the grain will become round shape easily and at the same time have many micro edges on the grain surface in large size fracture.
This paper describes the behavior of wire vibration during continuous machining with respect to the action of impulse based on a single discharge. The impulsive force acting on a wire is given by an experimental result of transient wave forms caused by a single discharge with the consideration of the simulation result supposing a vibration system to be a beam. The mean displacement of the wire obtained from the observation of continuous machining does not always obey the resultant displacement calculated by superposition of impulses from a set of single discharges. The vibration model is studied using a nonlinear model in which the number of impulses acting on the wire depends on the instantaneous gap distance between the wire and the workpiece. Thus, the behavior of the wire is significantly affected by the amplitude of the wire due to its tension or machining feed rate and by the discharge interval.
A thermal stress field induced by a line heat source often causes crack extension. Taking advantage of this phenomenon, some glassy materials can be cleaved without leaving any microcracks on the cleaved surface. In the present paper, the time dependent stress intensity factor of a line crack pre-introduced from the end of a thin plate has been obtained when a thin plate is heated by a continual line heat source on a crack. The stress intensity factor is affected by the width of the strip and the ligament length but is almost independent from the crack length. The present analysis agrees well with the experimental results using glass plates. The section of the glass cleaved by a line heat source was extremely smooth. The surface roughness was within 0.02μm at Ra.